[MUSIC PLAYING] This is Pragmatic, a weekly discussion show contemplating the practical application of technology. Exploring the real world trade-offs, we look at how great ideas are transformed into products and services that can change our lives. Nothing is as simple as it seems. I'm Ben Alexander and my co-host is John Chidjy. Hey John, how you doing? - I am very well, thank you, Ben. How you doing? - I'm doing well. It's a beautiful Ohio morning. - Well, it's, yes, I'd say it's a lovely summer evening here but in any case, okay, I'd like to start by saying some thank yous to several people that have said some very nice things about the show so far. I apologize if I mispronounce your name. Jamie Ryan, of course, I have that name down, Pat. Thank you, Jamie. Diego Petrucci, I think it is. Thank you very much. And also Brad Fortin, I think it is. And an extra special thank you to Mike Oertli. I think it is Oertli. So sorry if I mangled that. And he wrote a lovely piece on his blog about the show on his site. And that site, if you're curious, it has some interesting stuff there, is O-P-T-E dot C-H. So imagine Optech, but with a dot separating the C-H at the end. All of the downloads that we've been getting have really put Pragmatic essentially higher up in the iTunes charts than I actually thought would happen early days. But we're currently sitting number one, top podcasts in technology gadgets and number five in top podcasts for technology. They're also showing us on the front page as well when you first log in and look at any podcasts as a new and noteworthy. So thank you to everybody who has downloaded the show and who has listened to it. We really appreciate it. And thank you so much. If you feel like leaving a review, feel free. But in any case, I just wanted to say thank you before we got started. - Yeah, same here. It's been a wild past couple of days here. - Yeah, it has been. And without further ado, onto the main topic for this episode. And the main topic for this episode is automation. So, automation is essentially going really simplistic here. You want to be able to free up people from performing certain tasks. the purpose of automation is that you can get a machine to do it, you can build a machine to take over a manual task. So there's really two pieces you got to break it down into. You got to break it down into the devices that actuate and that is to say they move and manipulate the real world. And then you've got the controller. So need to talk about them sort of separately and a little bit of a history lesson as well, just to sort of get everyone's head in the right space. So first of all, I think in terms of digital controls and so on and digital communications, the simplest, most fundamental controlling device is actually a relay. And I'm not talking about a relay race, of course, no. I'm talking about a relay. And a relay, I'm just trying to think how you describe a relay. It's essentially an electromagnet. So when you run current through a coil, it creates a magnetic field and you do all of those coils in the same direction, bunch them all up together, you get a stronger magnetic field. So if you've got a ferrous material, that ferromagnetic, I should say, material, and what you can then do is you can run a current through the coil and it can physically move a switch. Same concept behind a solenoid, essentially a relay is a solenoid. But where a relay is different is that the thing it's switching is in itself another switch. Like a light switch, but without the physical button on it. So it can turn things on and off. So for example, you'll have a common connection and you'll have a normally open and a normally closed connection. And when you throw the relay, you power the relay up, it'll switch from one to the other. So in its non-energized state, there's no current flowing through the coil, the magnetic field doesn't exist, the spring loading will return the relay to its normal position. So if it's a normally open relay, normally open contact, then it'll go from, the connection will be from the common wire across to the normally open. Once you drive current through the coil, it will pull that towards the coil and that will close it to the normally close contact. And that essentially is a relay. And the point of a relay is that you separate two circuits. So on the one side, on the coil side, you could have whatever voltage you wanted, 24, 12 volts, 6 volts, 240 volts, 110 volts, doesn't matter. AC coil, DC coil, different, you know, different sort of subtly different design, doesn't matter we talked about power also in episode 2 so the point is that whatever you got on on the input side on the coil side you can have whatever you want on the other side of it on the actual side you're switching so you can literally use a relay to convert a signal from 24 volts DC up to a thousand volts AC if you wanted to and that idea if it's not perhaps immediately obvious, but it was actually invented. You know how things sometimes are invented at approximately the same time in history? Right. And especially when you go back beyond the time at which we had, I guess, maybe very, very fast, ubiquitous communication. So right now, if I hit send on an email and you hit send on an email, they'll have a time stamp on it, it'll be relative to UTC, and you'll be able to tell exactly who sent what first, the path it took, and all that information. Okay, obviously, that was not always the case. So, you go back to the Wright brothers, and there's an argument over who invented the first powered airplane, and then you go back to who invented the first car. And back in those days, where communication was not like that at all, you'd hear a story, "Oh, some guy in some other country you probably haven't heard of or certainly haven't been to invented something just like what you did. Well, so was the case with the damn relay. So the two guys is Joseph Henry or Edward Davey. So they both had, they both claimed to have invented the relay in 1835. But in the end, you know, hey, does it really matter? Someone invented the relay and it was a great idea. It was originally developed and employed at great length, I would say, in the 1840s, which is the dawn of Morse code telegraph. So essentially what this did was it allowed someone at one end using a very low voltage to basically press down a key and that would close a circuit that would drive a relay which higher voltage and that high voltage which travel along the telegraph lines on the other end it would do the reverse process it would step that high voltage down to a lower voltage there'd be booster stations along the way that had interposing relays to give it a kick as it went along. And that essentially was the main driving, I suppose, it was one of those things. I'm, it's not, history does not illustrate conclusively which came first. But I imagine that it was more that the relays popularity, it was used in the telegraph, for Morse code for communication over long distances. And that drove the fast adoption and recognition of how useful the relay was. Funny thing about relays though, is that they really haven't changed a hell of a lot since then, they really haven't. But they are the fundamental building block of most of the control systems. And even today, relays are present in pretty much every control system that you would care to look at. Sometimes you can't see them, they're embedded in the equipment. Other times, they're a separate item you can literally pull out a blown one that's busted and replace it. Okay so that's relays. There's a couple of good links in the show notes. I encourage you to look at them if you want to know more about the history of the relay. So we're going to fast forward now. So that's enough about relays. Just a little footnote, the transistor was invented in 1947 and of course the transistor is a silicon version of that, where a very low voltage or current can modulate a much larger voltage or current but it does it without physically moving contacts. The problem with relays is of course that they're a mechanical device so the magnetic field isn't mechanical but because you're opening and closing the contacts that creates a problem and there's all sorts of problems with that and I don't want to go into the details but the bottom line is if it moves it will break. So one of the things that in when you do reliability engineering, you realize that anything that has a connector, anything that moves or can move, uh, will inevitably fail before the things that don't. So if I solder, sorry, a North American expression, if I solder a device onto it, I mean, sorry, I have to point that out. When I first went to work in North America and I said solder, they looked at me like I was like, what are you talking about? I pointed at the wave soldering machine and they said, oh, you mean solder. And I'm like, yes, I mean solder, sorry. - Yeah, that's a weird one. - I know, anyway, I don't get that. The L disappears 'cause it's spelled the same way. Anyway, all right, back on topic. So when you solder a component down onto a board, it is going to be far more reliable than something that sits in a socket. Yeah, and anyone that's built their own PC, you know, you go back for the longest time possible. What's been happening is a rationalization of these components. gone from being socketized to all-in-one motherboards. To the point at which these days where you buy a motherboard for a PC, yeah, it's got a couple of expansion slots like PCI Express slots on it or whatever. And you don't have to put anything in it because the graphics card's on board and everything else is on board and CPU's the only thing really in a socket. And that socket's gone away from pins. It's gone to high pressure contact onto a flat bit of copper. So a lot of that movement has sort of taken up, whereas in the old days it was a pin alignment issue and everything. So moving parts anyway in the relay, I'm getting sidetracked. The moving parts in the relay are bad. So the transistor is significant because what it means is that you didn't have moving parts anymore. So that didn't start to have much of a dent on control systems until we get to about to the mid 60s. But it's still important to note. Okay, so, analog controls. I don't really want to go into too much depth on analog controls other than to state that they existed and some of them are still around. And what essentially analog controls are is operational amplifiers are, I don't know how deep to go on this one. Have a look in the notes if you want to know more about it, but essentially what you can do is consider an analog amplifier that will amplify anything that you give it into it in voltage from say 0 to 10. It'll output in linear response to that based on the gain that you set, and you can set the gain using different resistances. So the idea was that you had a whole bunch of these operational amplifiers and you would literally wire between them the resistances you wanted to quote-unquote "program" the controller. And all of the control parameters you would go through all the mathematics and you know, do all your Laplace transforms and figure out exactly how much gain you needed in order to control what you're trying to control. And analog controllers for example, would drive something like a PID controller. And that's all I want to say about analog controllers. Because PID controllers is really one of the big things in controls engineering. PID stands for Proportional Integral Derivative. And what essentially a PID controller does is let's say you're trying to control the flow through a pipe. And for whatever reason, there's a bunch mechanical reasons I won't go into where you have to control that to I don't know 30 liters a second you've got a pump on one end and you've got a water source and you've got to pump this up a hill you can't exceed 30 liters a second so you put a device along there to measure the flow a flow meter doesn't matter what kind just a flow meter and it says okay I'm now getting 29 liters a second. So what do you do? You run that through, that's your process variable from the field. You run the process variable back into a PID controller and you give it your set point. And the set point is 30 litres a second. It measures the difference between them and that is the error. And then based on the gain, integral and derivative times, it will then attempt to close that error by the output. And the output drives the speed of the pump. So the idea is that you modulate your speed of the pump up and down until you reach the desired flow rate. If that all sort of made sense, then PID controllers are extremely valuable because that's one example. Another example that I could think of, certainly back in the day when it was being developed, was positioning controls. So let's say you're on a battleship. And the reason I wanted to bring this one up is because I- a few years ago I saw- what was the name of the movie where um these aliens came- came down to earth and they had to get this old battleship out. They were all stuck in some kind of a big energy thing. I think it was just called battleship wasn't it? Yeah it was based on the- the board game wasn't it? I don't know something like that but the- the point is that they got this old- it was set around Hawaii and they got this old battleship from the 40s or 50s out and it had all of this, all the different, the old style controllers in there for positioning the gun turrets and the elevations and everything and I was looking at that thing and inside me I'm like, oh that looks so cool because that's all the stuff of course that doesn't exist anymore. It's all been mothballed and doesn't, you know, we've all gone to digital control and all that analog stuff is all gone But things like positioning controllers and all this sort of feedback and everything used to be handled with analog controllers and analog electronics, but everything's all gone digital because digital is so much better in a lot of ways. But at the same time, it's a hell of a lot more complicated. So just another silly example. I haven't worked on a battleship of that age, but looking at the controls on that and doing a little bit of reading about it, Yeah, a lot of similar concepts. So it's all about, you've got a variable that you're trying to control. You have a means by which you can affect that control. So by affecting that value and the PID controller allows you to set a set point and control it without having to keep your eye on it. Because of course, before analog controllers, they used to have somebody that did that. So what they would do is they would sit there with the speed control on the pump and they would simply say, "Faster, slower." And they would say, "Faster." They'd wait a minute they'd see the flow feedback coming back, oh, it's about 29.5 liters a second, okay, I'm a little bit faster. So it was all manually done, right? But this allowed people to get a external, well, essentially a mini sort of computer to do this, customized specifically though for that task. So that was just briefly analog controls, and they're very popular in the 50s and 60s, and all of those machines are more or less gone now. They're just not in use anymore. I think that they still had one at the university when I was there just for teaching purposes when we were doing operational amplifiers and control system theory, but we never even used it. It was like, oh yeah, that's an old analog controller. And we're like, oh, okay, that looks quaint, lovely. Next. So anyway. It was the USS Missouri. Was the-- - Ah, thank you, yes. - And which was where General MacArthur and Japanese Foreign Minister Mamoru Shigemitsu signed the surrender for World War II. - Yeah, yeah, that's it. - And it's a cool looking boat. - Oh, absolutely. Oh yeah. I mean, I'm not much into the military gear, but I can appreciate a nice looking boat like that. So that was just, yeah, I don't know. It's very cool. Anyhow, okay. So the next thing, the next thing that was happening at the same time was people were exploring, experimenting more with relays in terms of control. So what they were starting to do is they were starting to add things like timers and time delays into relays. So instead of just saying, I run a current through the coil, the coil pulls in the contacts and away we go. We've switched our current through from one to the other and it's gonna do whatever it's gonna do. Well, they start putting time delays in them. So you would, when you put the actual power on to the coil, it would then have a time delay that you could configure. It could be a second, two seconds, 10 seconds, whatever you want. And of course they then started developing ones did the reverse, which is once you take the power off, that it continues to maintain that contact for a certain amount of time before it releases the contact. And I would call those on delay and off delay timers. So with a combination of on and off delay timers and standard relays, you could start to actually build a panel with push buttons, indicator lights, that could start to do basic digital control. How exactly that works, It's very hard for me to describe. I've thought very hard about how I would describe it just using words because this is a, this is not a visual podcast, not a, what is the, what is the right name for that? A video cast? A vlog? Vlog. Yeah, something like that. It's this, that is not what this is. Vodcast. Vodcast. So I, I don't, it's really hard for me to do it without drawing something or showing you So please have a look in the show notes. There are a few good links in there. I strongly encourage you to have a look at those in your spare time to sort of give you an idea of how it would work. But the basic concept, if I can, I'll try and paint this single, very simple idea of a latching circuit, and then I'll move on. Consider you have a push button, and that push button essentially turns on the electricity. It starts from the source voltage or whatever the voltage is. So it's 24 volts DC. And I keep saying 24 volts DC. Just if you're curious, 24 volts DC has become the de facto industry standard for automation systems, and certainly in industrially. In telemetry systems, it's typically 12 volts because 12 volts is a standard car battery voltage. And 12 volt cells are typically used to backup for solar systems. So that's one of the reasons why 12 volts is predominant in telemetry systems that don't generally have a connection to the grid. But when it comes to industrial, it's 24 volts, which is not coincidentally twice the 12 volts. Anyway, bottom line, I keep talking about 24 volts. So let's just assume it's all 24 volts DC, which is pretty well standard in industry these days. You push the button, 24 volts gets to the next part of the circuit, which drives a relay. The thing is that those relays can have more than one set of contacts. So rather than just pulling one switch, you could pull two or three or four. So that's when we start talking about single pole relays, double pole relays and quad pole relays. Three poles exist, but they're pretty uncommon. It's usually one, two or four, or even eight, actually. Anyhow, so what we do is we will then run back to the start button and we'll say, you know what? Once you've pushed that button and closed the relay, I'm gonna feed back one of the outputs back into the input. And what that's gonna do is that's gonna keep the current flowing. So once I take my finger off the push button, the relay stays energized. And that's what they call a latch. So the circuit will then latch in and it will stay latched. And if you want to stop it, you have a stop button and the stop button is normally closed, which means the circuit will be complete if you don't touch it. And if someone says, oops, I got to stop this machine, I'm going to push the stop button, that'll break the circuit. Once the voltage disappears, the feedback disappears, which means the relay drops out. Hey, presto, you've got to start in the stop system. It's pretty simple when I draw it. it's real hard to explain that using words. So that's the best I could do. I hope that's okay. - No, I got it. It makes sense. - So the idea is now picture a plant that's got dozens or hundreds even of drives in it, motors, because of course the relay can drive big, big relays and the big relays, they call them contactors and a contactor can sink 20, 30 amps of current and that'll drive big electric motors to turn whatever it needs to turn or make conveyor belts move or whatever the hell they're doing, start and stop the air conditioning. It could be anything. The point is that you ended up with these massive, big cubicles full of relays, and you'd close the door on the bunch, there'd be a whole, on the front, there'd be a bunch of buttons, and they'd all be labeled, and that was your control system, was just an enormous cubicle full of relays. And I remember I saw one of these in operation when I was working at one of my previous companies, I was doing some service work with one of the companies that moved their facility up from Melbourne to Brisbane for tax reasons. So, the Queensland government was giving them tax breaks and all that sort of thing that, you know, they're trying to attract business. Sometimes cities temporarily go insane and do that sort of thing. And then they jack the prices up later and, you know, ha, you moved, too late now sort of thing happens and it's all very nasty. But anyway, I was listening to the panel, the relay panel, because the funny thing was we had a PLC and running the PLCs around the rest of the equipment, but the old duct work in the top, all the switching of the duct work was all done using relays and these relay panels that were built in the '60s. And I put my ear up to the cabinet and I listened to this song of all these relays going, click, click, click, click, click, click, click, click, click, click, click, click. And I'm like, that is just the most bizarre sound. To this day, it's imprinted in my brain. is just the most unusual thing because PLCs generally don't do that. The lights will flash on and off, but that's okay. You're not flicking relays every two, three relays a second. Because like I said, it's a problem, right? Because relays are a mechanical item and they will wear out. And you buy a relay, it lasts for 100,000 operations. Once those 100,000 operations are up, you're on your own. I mean, this thing might just stop actuating. It might get, one of the common failure mechanisms for relays is welded contacts or carbonized contacts. In other words, it's either always making a connection, no matter if you turn it off, or it'll never make a connection even if you turn it on. In which case, you know, you gotta chuck it out, put a new one in. - Well, and I'm sure it's, you know, if you're building a plant or you're building a, you know, a relay room, right? Then you've got your, there might be rated 100,000, but who knows, you get the bad one and click, click, click, **Matt Stauffer:** Right? It's just ... And I'm assuming that where that really becomes a problem is dealing with ... If something's relatively slow, you're not going to have as much heat and pressure building up, or as many fractures or tears in the material. But once you get to something really computationally intensive, it's going to start ... **Matt Stauffer:** Yeah, absolutely. **Matt Stauffer:** It's just going to run away on you. - Yep, spot on, exactly right. And the high temperature environments kill them and it's the number of operations that will kill it the quickest. But obviously if you're switching a relay 10 times a second and it's not gonna last a year, you know, and it's just, it's bad design. If that's what you've got to do with your control system, even with a PLC, you've done something wrong. You know, there's a better way of doing it. another way of doing it it's been done wrong. Relays aren't meant to be operated like that. So anyway okay here we are in the 60s we've got some analog controllers we're gonna do some PID control with we've got all these relays for all the digital control we need to do and then of course the transistor's now been developed to a point at which you can now have essentially mini computers or you had super computers they called them super computers oh god multiple room bunches of tape drives and anyway never mind that the point i'm getting at is that that silicon had started to come of age. So in the 60s, late 60s, it was 1968, there was essentially a group of companies that got together to try and solve this problem. And Bedford Associates, who was actually formed from a bunch of companies of Morley, Greenberg, actually these are founding members, I believe, Morley, Greenberg, Landau, Schwenk, and, oh goodness, Boissevain. I probably messed that one up as well, mangled, apologies. And they essentially formed Bedford Associates. And Bedford Associates came up with a design that they called the Modicon, which was short, which is an abbreviation for modular digital controller. And the model number was 084. And the Modicon PLC is the, essentially considered to be the first PLC. And a PLC is a programmable logic controller. And the whole point of a PLC, at least back at that point, was to get rid of the relays. It was to get rid of all the times and all that and to program it. So rather than actually, 'cause if you wanted to program a production line and all that back in those days in the '60s, you'd have to insert relays here, bypass relays there, mess with timers, and so on. All these sorts of electrical wires would run everywhere. One thing I didn't mention is I opened up that relay panel and had a look inside and it was just a dog's breakfast of wires running everywhere between thousands of relays. And I just looked at it and shook my head. I'm like, how could anyone understand this? And one of the older guys who'd actually brought this up from Melbourne had looked and he said, "Oh yeah, well this relay's doing that, one's doing that, and that one's doing that." And I'm just like, "What?" Yeah. Didn't they used to use relays to store memory as memory in early computers? Essentially, yeah. I mean, essentially when I described the latching circuit, a latching circuit is memory. So, a series of latched circuits, you could create a NAND gate, you could create, and from a NAND gate, you could create any form of gate, you know, so like AND, OR, NOT, NAND, blah, blah, blah. The point is that, yeah, you're right, they did. And thank goodness they came up with a PLC because suddenly it was possible with a form of software for you to programmatically change what the system would do. So you could change the time delays. You could just, you could add a time delay where there was one, where was, there wasn't one before just by adding an extra line of code. And that was a massive breakthrough. And the first industry to truly take that up was the automotive industry. So because I had big production lines and they were like, okay, we need to automate whatever we need to automate, like the assembly of this section, spray booths, I don't know, whatever they were automating, conveyor systems, but they were the first to really embrace the PLC. And the great thing about it was that there was this, in the next 10 years, there was a massive, let's all build a PLC, this is a fantastic idea. So there were hundreds and hundreds of PLCs that were brought to market in the next decade in maybe the next two decades. And of course, guess what? They were all totally different, different software, different ways of programming in different languages, different everything. But the one thing that they did keep, the one thing that still stands today is a common programming language that they call ladder logic. And the first time I saw ladder logic, it's sometimes called relay logic, but the first time I saw ladder logic, I sort of thought to myself, This is the most bizarre way of writing. Why? What? And the name comes from the fact that if you were to draw the relays, they had an abbreviated way of drawing the relays in these old relay control panels. And that relay logic diagram, if you drew a line connecting all of the coils on the right hand side and all of the source voltages on the left hand side and with all of the relay contacts in between. It looked like rungs on a ladder, like, you know, your logic literally looked like the rungs on a ladder that you would climb up to get into your gutters and your roof, and they called it ladder logic, and the name stuck. So, the Monarchons evolved, and a whole bunch of other competitors did too, but they all supported ladder logic. And initially, of course, it was very crude. The descriptions for how you would program was terrible. The great news was while all this was happening, computers were happening as well. So PLCs really, they are like a micro controller, a micro computer, but they're not, they don't have a keyboard, they don't have a mouse, they don't have a monitor, all they care about is input turns on, what do I do with it? I wait 10 seconds and turn the output on or something like that. So as time went on and computers developed, the software you could use to program them got better and better again, to the point at which it became visual. So you could literally drag and drop. So you want this, I want a contact here, I want a coil there, I want a time delay block there. You just drag it in, drag and drop, drag and drop, and you draw lines using a mouse. And that's sort of essentially where it's been for the last maybe 15-20 years. It's it's all become a visual, well, on the good PLCs anyway. The more expensive ones, yeah, they've got decent software packages because now everyone had computers and computers could be, were easily accessible. Because when PLCs first came out, it was all custom everything, custom controllers, custom programmers, custom everything. Whereas it had to die, it couldn't go on like that. So as a lot of industries, there's, when there's an innovation, suddenly everyone wants in. So every man is dog and, well, God, funny I say that, it sounds really gender biased when I say that. How about I just say everybody, everybody wants to have a PLC. So I can make a PLC better than everyone else. It kind of makes me think a little about like smartphones with touchscreens and smartwatches. It's like, oh, smartwatches, they're going to be the next big thing. Suddenly there's all these companies working on smartwatches. But the funny thing is that it's not always the first one to market that actually has the best product. And you need no more evidence than just looking at what Apple did with the iPhone as proof of that. They weren't the first smartphone in the world. No, of course they weren't. They were very late. When it came out, the idea was, I mean, the narrative was that they were too late, right? It had formed up. Yeah, that's right. Not true because they had had a superior product and similar thing has happened in the PLC industry. So PLC started out with so many manufacturers. The first one was Bedford Associates with Modicon and the Modicon PLC's changed hands a lot of times and now they're owned by a company called Schneider Electric. So the three big names and the biggest in the world in control systems is Siemens. The next one after that is a company that these days they call themselves Rockwell Automation. Most people however recognize Rockwell not by that name but by the name of the company that they acquired, Allen Bradley. So Allen Bradley is an American company that did a lot of work early on and Modicon was eventually went over to Europe and was developed further in Europe and now is owned by a French company and Siemens has always been Siemens. The gear that they've used, the most popular PLCs that they have, the current range they call them the S7, so the S7 300s and 400 PLCs, they have always been developed. They were based on the S5 before it and before that I'm... yeah, anyway. The point is that it's always been a German product by a German company and they started out essentially later to the game by several years but they've come out as being the most popular PLC in the world. There's a whole bunch of reasons for that but in any case those are your three front runners these days. So if you're thinking PLCs you're thinking of the big three. So when I say the big three I'm talking about Schneider, Rockwell and Siemens. Not in that order. Okay, but that wasn't the only problem because PLCs had other issues because when you have a production line, you don't have everything all in one place. What you've got is you've got control stations around the plant. So let's say, you know, I keep coming back to cars and the main reason is because well, car plants are highly automated and they're the ones that originally drove the adoption of PLCs in the first place. So anyhow, your spray booth is one of the last things you do. There's a whole bunch of controls associated with that. But then you've got, let's say you've got the side panel door assemblies, or you've got welding robots that are handling the spot welds on the trunk or whatever. All of that will have, there'll be different welding stations, there'll be different attachment points for when they put the wheels on, when they're fitting out the interior, and there'll be a big long moving production line. So you can't really have one PLC in one location and economically run cables out, wires out to every single actuator and every single feedback device that you've got, every button and back to one massive box. So what they had to do was they had to, not just for that reason, of course, there's also back in the 60s and 70s, there were constraints of memory because memory was precious. So there was only so much code you could put into one of these things, but it was just getting too expensive. So the PLCs had to be broken down and distributed. So you'd have one PLC in the spray booth, you'd have one PLC at a welding station and so on, and you'd break it down into multiple PLCs. And the next PLC would need to know a little bit about what the other one was doing. Hence, then we start with the protocol wars. And the protocol wars in PLCs were pretty vicious because everyone had their own way of doing it. It started out with serial and then it went on to, you know, packet-based data, essentially ethernet. And everyone had to do it just a little bit differently, didn't they? But the most common standard that's still used today that was present in the earlier, not the first modicon, of course, the 84, but in subsequent modicons after that was again lending from the name Modicon is called Mod Bus. So Modicon Bus, so Mod Bus. And Mod Bus has become essentially the de facto standard. If you have a device that has serial communication, then it should support Mod Bus because everything has the last 30 years. However, Siemens pushed their own ProfiBus, which is extremely very, very popular as well and Alan Bradley pushed DF+ a few other different ones along the way. But there are a hell of a lot more than that and thankfully most of those wars are over now and they've settled on those big three. And again not surprisingly they're attached to the three big manufacturers so Siemens pushed Profibus, Schneider pushed Modbus and Alan Bradley pushed the latest iteration of you know DF+ or ControlNet or whatever. So that was a serial communications and that was happening basically late 70s and into the 80s because you got to realize at that point that the internet didn't exist. Networking and Ethernet was not very common at all and the whole DARPAnet thing and all that, that was all happening off in another world essentially from automation. Ethernet really didn't start coming into products until the 90s and even then probably mid 90s for some of them. I think Siemens started to adopt it in the late 90s. Started as a separate add-on module and then when you get to the probably about 2002-2003 they started to incorporate it directly into every one of their CPUs. Like now you get Ethernet now. It's like yay! Hallelujah! Much better. Okay so all the Ethernet stuff they came up with their own version of it. So you have Profinet for Siemens and ControlNet for for Alan Bradley stuff and there's device that as well in there. And Modbus of course Schneider decided to go with Modbus TCP because well, it's a brand. So, let's just call it Modbus but with TCP, which is of course essentially the same thing but it works over a packeting system. Okay, I think that's enough about comms and enough about PLCs 'cause there's two other really important things that have happened in industrial automation. And then we'll start talking about the other niches of automation like building controls and of course, home automation. Okay. So the next big thing to happen, as I was saying, the development of computers is happening at the same time. And that was good because it drove not only the programming software, but it also drove the monitoring software. So it's one thing to program a PLC to say, turn this on, turn that off, time to lay here, blah, blah, blah. would love to do is you would love to have it all displayed on a computer screen and you could look at it from 100 meters away or 100 miles away and see what was open, what was closed, what was running and what was stopped. And of course these things have faults. Well bring the fault feedback into the PLC as well. Let's display that. I want to see if there's a fault on any of my lines. Because then what you can do is you can actually completely remove people from the environment. and you can say, "Right, you sit in the control room, "you sit in front of this computer screen, "and it'll show you the state of everything "that's happening in the entire plant." And that has massive cost-saving benefits. Of course, the flip side of that is you're putting a lot of people out of work, but that's a moral discussion. It's worthy of discussion, and I don't mean to dismiss it, and I don't like to dismiss it because it bugs me a lot, but I just don't want to talk about this just today. So rain check on that discussion. Okay, so they came up with a obscure name for this stuff, which they always seem to do. It's called Supervisory Control and Data Acquisition. And that stands for SCADA. So when people talk about SCADA systems, they're talking about software that's specifically designed to take data out of the PLCs and display it visually for someone to either observe or to interact with. Because what you could do is, rather than having a physical push button now, a PLC having a memory, you could turn it into a virtual push button. So if I click on a button on the screen, it has the same effect now as physically pushing a start button for that device in the field. And that gives you not only monitoring, but control. And that's very, very powerful. So as computers have developed, that's gotten better and better and better. And it's just reached the point now where SCADA is immensely powerful. And SCADA did the same kind of thing that everything else had, which was there was a massive rush of people to the party. People that had PLCs felt that they had to develop their own SCADA to go with it. And industry standards weren't very, there just weren't very many. So Siemens developed what they call WinCC and Allen Bradley, they were developing RSVue. and Schneider developed Vigio. But there are other companies that developed SCADA packages that were designed specifically to be generic. And one of those is an Australian example that was Sitect and Sitect is very popular because originally supported over a hundred PLCs. So if you wanted to talk to a Siemens PLC or Schneider PLC, different PLCs that wouldn't ordinarily talk happily to each other, they would happily talk to the same SCADA system. So you could literally have a production line made of different PLCs and connect them up to one Scala system and away you go. So that had its attractions as well. But to solve the problem, they came out with something called OLE for process control, object link, object linked. I always forget what that stands for, object linked embedding, I think it's something like that. And that's, OLE for process controls, short for OPC, and OPC is what everyone calls it. And OPC is essentially, I'll give you an OPC driver for my PLC, And that means that any SCADA system that supports OPC can talk to it. So there was no longer a need for Scitectin to exist anymore, and you could use whatever SCADA software you wanted, as long as it was OPC compliant and you had an OPC driver, you could talk to whatever PLC you wanted. And that all happened in the late '90s and in the 2000s. So, hopefully you're still with us. We're almost there. And that is the DCS. Some people talk about SCADA PLC, and then they talk about DCS. A DCS is different from a PLC and SCADA system. It was developed later historically, because what it is, it's a specifically designed integrated system. In other words, you have terminals and you have PLCs, the two are linked. You both modify the code on the DCS and you display the results on the DCS as one integrated solution. And that has a whole bunch of advantages and it has a whole bunch of disadvantages. So rather than go into either or, I'm just going to leave it there. So, you know, DCSs are, they're kind of like PLC plus SCADA. Okay. I talked a little bit about programming languages. Just wanted to cover the famous IEC 61131 standard. So the international standards body said it kind of sucks how you've got all these PLCs, all these different programming languages. Sure. They still support ladder logic, but you know what? I can't just export code from one PLC into another. You know, I just, I can't do it. So what we want to do is we want to standardize in programming languages. So they came up with that standard with 61131. And it spells out what it should look like, how it should work, and then manufacturers would say, right, our code conforms with this. So yay. And it specifies a whole bunch of different languages, not just ladder logic, something else called functional block diagram, structured text language, There's a few other different ones as well in there. And the whole point is standardization, which has been really good for the industry. Well, no, that's not fair. It's been really good for the programmers. Whether or not Siemens are happy about it, I'm not so sure. You can tell just by the names of these products that these companies are shipping that they're not too keen on giving up that kind of control. No, no, no, no, definitely not. So, because there's no incentive for them to do it, you know, none at all. Because what they want is they want you to say, right, well, you've got a Siemens PLC because, I mean, let's face it, Siemens PLCs are pretty great. However, I don't like their SCADA. Their SCADA WinCC, yeah, it's passable. But, you know, I prefer to do something like Clear SCADA or, you know, Honeywell Experian or something like that, you know, whatever. imagine if it didn't exist and you had no choice, you had to get WinCC, otherwise you couldn't talk to your Siemens PLC. That'd be terrible. Then they could ship you 10 consultants. Yeah, you're getting it. That's the problem, right? Okay, so we've talked about the history of automation, all the different booms and all all the consolidations, the key players, the different kinds of equipment. One of the things that's become more prevalent lately is the PLCs are now starting to support redundancy. So CPUs with power supplies being high failure rate items. So now a lot of the push in PLC land is for redundant systems. And another one of the pushes lately as well has been safety systems. So the idea is that you can't, It's not possible, for example, to defeat a series of interlocks if it's a protective device. And if that doesn't make too much sense, then, well, I probably shouldn't go into that, actually. But bottom line is that safety systems are now becoming a big thing as well. And I guess at that point, I just want to push pause on industrial automation and start talking about where it's used. Because people don't realize this, that automation is used everywhere. However, what's happened is that certain niche markets have sort of evolved into specific use cases. So a specific use case that I can think of that's very common are railways. So railway control systems, they use PLCs, they use SCADA as well, but what they do is the way they display their information, some of the protocols that they use have been customized and in some ways I'd say refined a little bit, but they're more specific to railway signaling and switching points and all that sort of stuff, because they measure things like occupancy on train lines and so on. So, you know, because when a train is sitting on train tracks, it's creating a conductive path between the two. You see a short between the two tracks, then you can tell if there's a train present. And all that sort of stuff is subtly different and unique to the rail system. It's not something that you would find at a wastewater treatment plant or a water pipeline. However, there's other examples like heating and ventilation systems. So they have their own controllers and with all the thermostats feeding into the controllers to drive the air conditioning system, the compressors and so on, chillers on larger systems and all that stuff. And another one is just on buildings is building controls. So things like door locks, solenoid door locks, fingerprint scanner access in and out of certain areas. Motion detectors, lift controls is another example of a control system that's very niche. And one time I was having a look at the top floor, on the top floor usually, but not always, the controllers for a building I used to work in. And I was fortunate that I came up when the lift guy from Schindler Lifts was there and he had the panel open. And of course, I couldn't help myself. So, I'm shoving my head in there and I'm asking him a hundred questions. He's looking at me and I'm like, I'm sorry, I'm an engineer. And he's like, oh, I should have guessed I was in a building full of engineers that someone was going to say something. I'm sorry, I'm sorry. I just couldn't help it. I'm drawn to the flashing lights. Anyway, but yeah, I mean, it was a customized controller, but essentially, he was programming it much the same way that you would program a PLC. had software that was built for it, but a lot of the same concepts applied. It's just that it had been fine-tuned specifically for that application. So a lot of the PLCs that I've been talking about, they're essentially a generic PLC, a generic controller that you can apply to all sorts of scenarios to do digital control and analog controls. So PID controls can now be done in a digital PLC. You don't need an analog controller anymore, but it's also generic. So you could use a PLC to control a lift, but it's gonna cost you more because it's not specialized for controlling that lift. So you're far better off using a customized controller for that lift, but it's still technically a control system. And now we get to home automation. Home automation is yet another example of a control system, but it has been essentially, I don't want to say dumbed down, but simplified in its scope, into what it is insofar as what it can do. And honestly, I've looked into this a little bit, but the thing that I've looked into the most that I'm happy to talk about is lighting controls. Not so long ago, and when I say not so long ago, I mean like maybe 10 years, it was the case where if you wanted to have controlled lights you needed to, hang on, I need to step back a second. A standard light switch is literally a physical toggle. You flick it on and off. You've got the slider type ones where you slide them up and down, and then you've got the ones with the little knob that flicks out. You literally flick it up, flick it down, and all it's doing is it's switching from completing a circuit to breaking a circuit. Pretty straightforward. But what you need if you want to control that is you need to go from a physical switch to a digital switch. And the way a digital switch works is that they're the ones that are literally like a little push button. And they come in all sorts of different forms but a little push button about the same size as your standard light switch. And that digital push button will sometimes have a feedback LED on it, like a blue, green, red, whatever colour LED, to tell you that it's on. you if you look up the light, you notice the lights on. So, you know, you can figure out it's on. But anyway, I just I wonder about that. Yeah, it's like it's a purr force, don't you think? It's the it's the the the the dream of a weather app that tells you if it's raining, right? Yeah. Problem solved. Yeah, well, that's yeah, shivvy head out the window. This is the thing because I think someone once made the argument to me when I was having this when I started on that little rant. And And they said, "Well, okay, what if the light bulb's blown?" I'm like, "Well, what are you going to do with a normal light switch? Forget it's digital. What are you going to do? You're going to turn the light switch on, then you're going to turn it off. You're going to turn it on, you're going to turn it off. You're going to say, 'Oh, hey, it's not, the light's not working.' So why do I need a blue light on the front of my digital light switch to tell me if the light is on or not? I'm just going to do the same thing. Push button, push button, push button. Oop, I'm not getting light. It's busted." You know what I mean? It's just, I don't know. - It's just the features benefits thing. - I'll say it looks futuristic and it looks cool, but it is pointless. - You have a little bank of it. - Yeah, exactly. But that's another thing I hate about blue LEDs at nighttime, right? 'Cause I have a couple of hard drives with those blue LEDs to show when there's... And they reflect around the, 'cause I've got light colored walls, the light reflects around and, you know, I can sort of see a very faint blue glow coming from the study of a night time. Anyway, I haven't got digital lights in this house. I'd probably make sure I'd got ones that didn't have the indicator LEDs on them. Anyway, I'm digressing again. The point is that all these are, are they a relay with a latch in them such that when you push the button, it closes the relay, latches itself shut, and that closes a circuit that drives the light bulb. want to push the button again then a digital controller detects oh it's already on therefore I'm gonna open my circuit breaker and drop out the latch um I said circuit breaker drop out my relay sorry and open that latch. The light bulb turns off. So why the hell would you want to do that and that's the reason you want to do it is because that opens up the possibility of automation. So I no longer have a physical switch turning it on and off I have a digital switch that I can turn on and off via a relay, it doesn't have to be actuated by a button locally. It could be actuated by a button from the next room, from essentially the equivalent of a SCADA system with a PLC. So there's the two pieces once again, just like I said before. You've got the controller that does all of the pulling the strings, the puppet master, if you like, if you're into analogies, and you've got all the actual actuating devices, a digital light switch, whether it's something that actuates and opens and closes the curtains or the blinds or the windows, or whether it's inputs for motion detection to tell when you're in a room or presence detection when you're in a room, when you've left the room. All of these things can feed into a controller and it can decide what to go do with itself. So about 10 years ago, circling back to that, the only way that was really available was you would have to run a separate set of cables to connect up all of these light switches onto a common communications bus. There were no wireless options. So what you had was the same problem with geeks that want to wire their house full of Cat5 or Cat6, I guess, these days. So Wi-Fi is awesome because that's what everyone likes to use, right? You put Wi-Fi in a room and you can use your, well, I guess back in the day it was laptops, but now it's everything. It's iPads, iPhones, and non-Apple products that have wifi. Whatever they may be, no judgment. But the point is that you can wander through the house freely. You don't have to jack into the wall. You don't, and the big reason is that's not a big deal if it's there. You know, I mean, okay, sure, I guess you've got cables and you get trip over cables and that's a pain in the neck, but you'll always get better transfer rates out of cable than you'll ever get out of wifi. you just, you will. The technology for wired is always gonna be ahead of wireless. And yeah, and wireless is a fixed resource, which means once it's gone, it's gone. Once you've used it, you can't use anymore, you're stuffed. Whereas with a cable, you just add another cable. The signal's contained within the cable. You add 10 cables, you can have 10 times the bandwidth. Eventually you run out of bandwidth on wifi. So there's always a limit. Anyway, so people didn't wire their houses with Cat5 is a pain in the neck. You know there's a lot of cabling to do and you've got to have you know an Ethernet switch connecting it all together to make it work. Well the thing is that these light switches are the same problem. So you had to run not only power cables to these things but you had to run a second bus around the house that didn't follow the power circuits that followed the actual light switches. And the most common standard bus that they went for is a is a sea bus, as in the letter C, sea bus, as opposed to a bus on the ocean, which the sea bus that they have in Vancouver, I guess that was confusing 'cause I was aware of sea bus and I went to Vancouver on a holiday once and I'm like, "Oh, it's a sea bus," Engineer Giggle. Right, I was the only one that got that joke, so we moved on, but anyway. So, okay, the controller itself had much simpler programming software, But the idea is simple. You can say, well, a light should not stay on for more than, I don't know, 10 minutes, I don't know. If you've got motion detections or presence detections in the house, some of the more advanced controllers can say, if no one is in that room for more than five minutes, turn the light off. Or if someone walks into a room, turn the lights on immediately. Yeah, even more fancy systems you can have, yeah, that they'll detect the light level automatically. Of course, people these days are more interested in the cool remote access stuff. So if you wire up your house for home automation and you've got a lot of these sensors, then it's not so much about the automating of it, it's just the remote control of it. So you can say, oh, I've got my iPhone and I'm gonna go and turn the lights off in the kitchen. Tap, and it's like, oh, that's the coolest thing ever. I mean, practically speaking, who cares? But hey, it's so damn cool. - Right, you think about like the Nest thermostat and what's really cool about it is that it's the thermostat that learns, right? That's the pitch, but you know that, gosh, what everyone really loves about it is that they have an app for it. - Oh, absolutely. And I sort of think-- - Even if they never use it. - Yeah, exactly. But you see that it's, what I find fascinating is that home automation has been around for quite a while now. It's been around for at least two decades. - Yeah, I think like all the old X11 stuff, right? Was that the company? - Yeah. - I mean, I remember a buddy of mine had that and he's like, "Check it out, I can, you know, I can, we can use our Kyocera 6035s to control the lights." And it worked. - Yep, that's right. But there were two problems and one was all the extra wiring you had to do 'cause it was fully wired controllers. It wasn't wireless. And that added a lot of expense because it was usually retrofitted into houses, which is a massive pain in the neck. But it's not just that, it was also that the controllers were simply, there weren't enough of them and therefore they were very expensive and it was still a very niche market. The thing is today, it hasn't changed a hell of a lot. It is still a very niche market, but the difference is that more people are doing it because now they've gone wireless. So there are now wireless equivalents of the CBUS standard and there's a whole bunch of different ones. And what effectively you can do now is you can circumvent the wiring problem. And it's at the point where you still need to be an electrician to go and plug these things into the actual switch on the face plate on the wall. Yeah, that's still a barrier to entry. So what's happening now is there's an even more interesting twist to the story, whereby they're building in the controller into the light bulb. So instead of having a switch you turn on and off on the wall that you can access on and off remotely or through some kind of control. If someone leaves the room, it turns off after five minutes, whatever it is. Instead of that, now you can remotely control them through the light bulb. And all you gotta do is screw the light bulb in. It gets power from when the light bulb's on. You always leave the light switch on at the wall. I mean, like at the face plate, it's always switched on. And then you can control it through an app on your phone, through the local wifi. And our mutual friend, James Smith. - Right, yeah. - Working for LIFX, I think it is, isn't it? - Yeah, that's it. Yeah, he's got the light bulb that, I mean, it's rated for, I don't know, I mean, years and years and years, very long life, right? And yeah, you just, you plug it in and or you screw it into the light and that's it. You can control it from anywhere. You can, it's pretty neat. got their, it's color control too. So you kind of have this new dimension to play with. I mean, and that's, to me, that's really interesting. I didn't see it coming. And I mean, this is just one example. There are dozens of others, but theirs has got all these other things you just mentioned, and this is not an advertisement for them, but, you know, by all means, check it out. It is kind of cool. What I mean to say is that the idea of, you know, "How do we make this more accessible? "How do we get this out to more people?" Because home automation should be a given. It should be up there with, if you live in a warm climate, you get a solar hot water system. If you live in a warm climate, which gets a lot of sun, you get solar hot water and solar panels. It should just be everyone gets home automation built in when they build a new house. It shouldn't be a massive expense. But it just wasn't happening because it was a little bit too expensive and it was just too much of a luxury thing. So how do you bypass that? Well, you take the electrician out of the equation. You don't need to run wires through the walls. You don't need to get an electrician to swap out the faceplate anymore. You can just buy a special light bulb or an adapter for an existing light bulb. And hey, presto, it'll power itself. You simply leave the switch on and then you can remotely control it through wireless means. And that's where everything seems to be going. And you can get exactly the same sort of adapters, not just for light bulbs, but also for standard, what they call a general purpose outlet or a general purpose socket outlet. There's all sorts of different names for it, I guess, but in Australia, that's what we call them. But that's the three pin power point in the wall. So now you can control an individual electronic device, whatever it may be. And I mean, up to a point, because what's the good of that on a TV set? 'Cause TV sets these days have got an auxiliary circuit that waits for someone to push a button. So sitting there idling, consuming electricity all the time, waiting for someone to come on and push the on button or push the magic button on the remote control, the infrared remote control to start the TV up. So what good is a remote control on that? You can turn it off, but you couldn't turn it back on again. And what's the point of remotely turning on a television anyway? It's not like it takes time to warm up anymore. It's not like it was back in the fifties where you had valves in the back of your TV set and you'd turn it on you need to wait five minutes for it to warm up before it would show a picture. - Right, now you just wait five minutes for the cable box to figure out what the hell it's doing. (laughing) - Yeah, that's true. That's true. So, yeah, I find home automation is going to take off more and more, but the thing is, I don't understand so much the attraction of the remote access. It's more of the automation side of it that's useful. So I went through this thing, and I just getting close to wrapping up is the situation I looked into it in was a cost trade-off with my kids. It may or may not make sense, but kids tend to be a little forgetful of things, and light is one of them. So, for example, it becomes a drill. Whenever we leave the house now, that I walk through all their bedrooms, turn off their light, we're in the middle of summer, turn off their fans as well. So our whole house is in air condition, we just have fans in the rooms. So I go around the house and do all that. Would it not be lovely to be able to push a button and say, "Shut down the house," and just turn off all the lights, all the fans, or go to a I'm out mode where you've got a key light turned on. 'Cause you know, some people like to say, oh, you should always leave a light on, you know, so it looks like your home, so burglars will be confused, right? - Right. - Yeah, like that works. But yeah, well, maybe it does, maybe it doesn't. I don't know, I doubt that as well. But nevermind, let's just run with that. Sure, leave a light on. Either way, you've got like a I'm away from home mode. Wouldn't that be awesome? But you see, that's not the angle that makes sense exactly. I mean, that's a convenience thing, but what makes sense is money, in terms of money for electricity. So the way I looked at it was, how much would I have to spend on an automation system to reduce the amount of electricity wasted by lights being left on unnecessarily? Obviously, you gotta make a whole bunch of assumptions, but it's really not as hard as you might think, because if you have a system that's set up such that you have a motion sensor in a room, someone leaves the room, no one's in the room for two minutes or one minute or 30 seconds even, you know, turn the damn light off. Yeah, what do you need it for? If you're sitting in front of the TV set, maybe you don't want the light on anyway. If you're sitting quietly in the study and you're not moving a hell of a lot, no, I guess it comes down to the sensitivity of the motion sensor, the presence detector. But the bottom line is that it should be possible to put that together. And if you can figure out an estimate of how much electricity is wasted. Well, I did the math on this place and the usage patterns that I witnessed with my kids. And I basically came to the conclusion the payback period was about 11 years. But that's still worth considering. So I guess what I mean to say by that is that the cost up front, it would cost me to retrofit all of these home automation devices to, and do the programming. Well, okay, I didn't cost my own time to do the programming, but nevermind that, that's okay. I'll work for free if it's my own family. But, you know, all of that is gonna end up costing me a fair bit of money, and I'm not gonna see benefit until 11 years have passed. So that's the sort of investment that is a harder sell. putting solar panels on the roof, the payback period was significantly less than that. And it was a bigger payback afterwards. That was worth doing. But your use case may vary. But that's just, that was my experience. And it's an angle that some people don't consider. And I think it's worth people considering with home automation. - Well, you know, talking about this made me think of Flux, the little app that we got it for Mac and I'm sure for PC too. And I think even some, like if you jailbreak your iPhone, I think you can put it on there too. And it's a little thing that warms up or warms down your screen's color balance as the day progresses. So you're not looking at this super bright, I don't know what 6,300K screen at 12 o'clock at night. And the idea with it is that you just, you probably shouldn't be doing that, right? You probably shouldn't be looking at things that are as bright as the midday sun when you're about to go to sleep. And it's, you've got, so you've got your cost, you've got kind of your all or nothing sort of, approach, right? Where you walk out of the room and 20 seconds later, the lights flick off. But you also, I mean, it enables these kind of these, these gradual things too, where you can, you know, and maybe flux doesn't matter. Maybe it's, you know, I'm not sure it's completely clear that the science is there, but I, you know, when I came down this morning to do the, uh, to do the show, the screensaver is running on my Mac, you know, and that's probably just cause I have it set. So it doesn't turn on right away. And I probably I just have something set wrong, but I'm thinking, you know, it's got a little camera there. It knows when it's being used. Why do I even have to bother with this at all? Right? Like, yeah, you should run the screensaver for a little bit, but after seven hours, I've been asleep and you know what time it is. And presumably this machine can also keep track. I mean, obviously it can. Know when I'm likely to be using it. That there's no good reason that that was still on. And granted there's very little power that's drawing out of that, but it's there. I mean, it's something and it's not great. So the more kind of smarts we get into everything, we can start enabling these sort of little, personal customizations, right? That, I mean, you're talking about with kids, not turning the lights on and off. I mean, what you're also doing there in addition to the cost is you're sidestepping the social cost, right, of having to deal with that, having to be the dad that's always yelling at him to turn off the lights, right? - Yeah, that's right. - And it's hard to put a number on that, but it's there. I mean, there's some aspect to that where it's a problem that we don't have to have anymore. - Yeah. See, the thing is, when I did that costing, I sort of, I made an offhand comment, I'll program it 'cause, you know, on my own time, I'm, you know, with my family, I'm happy to work for free for my family, But that sort of hides the other little sub-issue that you just hit on, which is putting a cost on that, the fact that I would no longer have to wander around the house to do that. So that wandering around the house, let's say I leave the house with the kids once a day, which is really not that unrealistic. In some cases, you come and go from the house multiple times per day, depending on kids' sports, birthday parties, family outings, events, school, it adds up more than you'd think. And so, now it takes me to wander around, turn all the lights off in the house and everything. Let's say that exercise takes me two minutes. Well, actually, let's say it takes three minutes. So, you do that just once a day, seven days a week. That's 21 minutes of your time. Over a year, that's going to add up to many hours of your time. Right. And one of the things that I often sort of think about is there's only so much life that each of us has. And I don't want to go all, I don't know, fluffy, but it's- I'm not meaning to be. It's if you want to put a cost on every hour that you spend and say, I want to do something that is going to generate revenue. And I don't say that that should be the point of life, but let's just say that for the argument's sake, that you want to look at it that way. You could say, "Well, I could spend those hours and hours of my life doing something that's going to earn me money, that's going to earn me like developing a website that makes money or writing code, writing software that's going to make me money." Or, honestly, if you want to take the money away from it, just something you'd rather be doing. Right. Yeah. I mean, it works better if you think about different currencies, right? That there's the- Yes. the social currency inside your family or with your friends. And then there's your own... What I just said this morning, I woke up and I said, "Why is the screensaver on?" And it was just a little thing, but it adds up. And all those experiences over a lifetime end up being some pretty big numbers. You have to decide how you want to count that or if you even care. And the flip side of it is there's the, maybe it's like the Protestant work ethic of like, "Well, you should be suffering. You should take the time to march through your house and do it." That can be… I mean, there's some value to that, right? That it builds character, that thinking about things and being thoughtful and remembering is good for you. But there's lots of ways you can do that. It's kind of an insidious mindset sometimes to think that there's somehow value in repeating yourself. There's the programming idea of keeping it dry. Don't repeat yourself. If you've written a little piece of code more than twice, it's time to encapsulate and pull it out and have it do its own thing. So, you don't have to bear that cognitive load again and again and again and again. And it's worth exploring. The answers might not always be the same. And a lot of times you're going to say, "Ah, it's fine." But you don't want to ignore it either. Yeah. I find that it's an interesting argument because what someone will say is, you say, "I build character." Man, I've heard that too many times in my life from other people. Honestly, it's a simple choice. Is there a better way or isn't there? If there isn't, you're stuck with it. As soon as there's a better way, you have a choice. And once you've got a choice, you can choose to either work- there's another one of those sayings, I'm about to spit it out. So, here it comes, which is, work smarter, not harder. Right. And I say that with a certain tiresomeness in my voice, because, God, I've also heard that so many times. Anyway. But, you know, irrespective of how worn out that expression is, it does have a good kernel of truth to it, which is, you should be looking for better ways of doing things. Because if you don't look for better ways of doing things or ways of saving time that you could put towards other endeavors, whether they are to make money or not, freeing up your own personal time, different currencies, as you were saying. That's worth exploring, accepting everything that we have and saying, "You know, Light Switch works fine." And go, "Yep, I've got that plenty of times." Actually, most recently from my wife when I proposed this. "I just got 11-year payback and Light Switch works fine." Anyway, I'm working on it. The point is that you should be looking for better ways because life is short. There is only a certain amount of time. I think that if you can use technology to enhance that aspect of your life and reduce your dependence on menial tasks, then surely that's got to be a better net result in the end. Right. And there's the other thing that goes along with that is when you, it's just, it's human nature a little bit that when something is, it's like doing the dishes, right? I mean, everyone knows the smartest way to do the dishes is as soon as you're done. As soon as you're done eating is to clean them off and get them in there so you don't have this, you know, pile of dishes piling up. But when something's a chore or when something has, you know, you're not, you're not getting any real immediate gratification or pleasure out of doing it, there's a good chance you're just going to let it slide. So if your option isn't to, it isn't just, "We're going to automate our lighting and we have two choices, which is to either do it or to not do it." You have three choices. It's to, you can automate your lighting or you can not automate your lighting and manually go and make sure that you're turning everything off, or you can not automate your lighting and just leave the lights running all the time. And if you, you know, if you think really honestly about your behavior and, and, you know, your kids' behavior and your, your family and friends and, and how we all act as humans, a lot of the time, we're just gonna leave the lights running and not even think about it. Well, that's a way worse outcome than automating it. So you have to think about how you behave when you're thinking about something rationally and being smart about it versus, you know, we're all dumb most of the time. And we, you know, making the smart you be really friendly to the dumb you. It's an interesting way of looking at it. I think the whole thing with home automation and what it can do to reduce some of the pressure on your life in different ways. The way I looked at it is I need to see a cost saving of either in money or in time. And in this case, it ends up being both, which is a win-win. And if I do spend the extra money, what's the payback period? I feel like there's a sort of a pervasive belief amongst people with home automation and they say, "Oh, home automation is wonderful. It's this, it's that. I can turn off the light bulb even when I'm on holidays in another country. And isn't that awesome?" And everyone sort of seems to look at that and say, "Well, that's nice, but I mean, that's what's the point of that? That doesn't, yeah, what value does that really give you?" And I think that the people that are in that situation are looking at it wrong. They're looking at it from the point of view of, "Just 'cause I can do something from the next room, that's not the power of the system." The power of the system is in the automation side of it, not the remote control of it. And that's the automation is what will save you the time and it will save you the money. And that's, I think, the angle that people need to consider it from. And I just don't see enough people doing that. And it is out there, it is becoming cheaper, and I strongly encourage people to start looking into that. - Cool. - Do you wanna... Wrap it up? Yep. So if you want to talk more about this, you can find Jon on Twitter at JonChidgy. That's J-O-H-N-C-H-I-D-G-E-Y. It's the same on app.net. If you'd like to send an email, you can send it to Jon at techdistortion.com. I'm Ben Alexander and you can reach me on Twitter @fiatluxfm or you can see show announcements and related materials by following the show account @pragmaticshow. Thanks for listening everyone. Thanks, John. Thank you. [music] Thank you. (dramatic music) [Music] [BLANK_AUDIO] [BLANK_AUDIO]